Routing and scheduling in multihop wireless networks with time-varying channels

  • Authors:
  • Matthew Andrews;Lisa Zhang

  • Affiliations:
  • Bell Laboratories, Murray Hill, NJ;Bell Laboratories, Murray Hill, NJ

  • Venue:
  • ACM Transactions on Algorithms (TALG)
  • Year:
  • 2007

Quantified Score

Hi-index 0.00

Visualization

Abstract

We study routing and scheduling in multihop wireless networks. When data is transmitted from its source node to its destination node it may go through other wireless nodes as intermediate hops. The data transmission is node constrained, that is, every node can transmit data to at most one neighboring node per time step. The transmission rates are time varying as a result of changing wireless channel conditions. In this article, we assume that data arrivals and transmission rates are governed by an adversary. The power of the adversary is limited by an admissibility condition which forbids the adversary from overloading any wireless node a priori. The node-constrained transmission and time-varying nature of the transmission rates make our model different from and harder than the standard adversarial queueing model which relates to wireline networks. For the case in which the adversary specifies the paths that the data must follow, we design scheduling algorithms that ensure network stability. These algorithms try to give priority to the data that is closest to its source node. However, at each time step only a subset of the data queued at a node is eligible for scheduling. One of our algorithms is fully distributed. For the case in which the adversary does not dictate the data paths, we show how to route data so that the admissibility condition is satisfied. We can then schedule data along the chosen paths using our stable scheduling algorithms.